Closed Loop Forecasting

This document explains the main principles of the M3 concept of closed loop forecasting.

The closed loop refers to the integrated forecasting and demand planning process in M3 Business Engine (M3 BE) and M3 Demand Planner (DMP). Forecasts, demand data and sales budgets from M3 BE are imported into DMP for further processing and then exported back to M3 BE. Using this concept enables greater precision in demand planning and forecasting, and facilitates coordination in multisite companies.

Outcome

Closed loop forecasting results in a daily gross forecast.

The forecast is used as a manually entered forecast within material planning in M3 BE. The results from the demand planning in DMP can also be used as input to M3 Supply Chain Planner.

During this process, the dataset file is updated when data is exported back to M3 BE from DMP. The user can decide whether to export all data or new and/or changed data only. If the dataset is transferred to item forecasts in the forecasting (FOR) component group, the forecasting file (MITAFO) is updated.

Before you start

  • The company must have a policy for how and when to perform collaborative demand planning using DMP.
  • The necessary enabling processes must have been performed. (Refer to the respective enabling processes for further information.)

Purpose

The purpose of using this concept is to refine the forecasting by making it possible to forecast close to the end customers and still maintain control over the extended supply chain. Several forecasting participants at different places can work simultaneously with their respective forecast areas.

When

How often closed loop forecasting should be performed depends on the company’s time fences and forecasting cycles. For example, a fashion company might perform it once per month, while a food company performs it every week.

How

Data that should be used in forecasting is accumulated in datasets, which are storage sites for budgets and statistical information in M3 BE. This data is then imported into M3 Demand Planner. An administrator creates DMP datasets and forecast packages and distributes these to forecasters that are located near or have further knowledge of the company’s different markets. The forecasters review and, if necessary, change the values or create new ones. The administrator then consolidates these values into an updated sales budget. This budget is exported back to M3 BE, where it can be transferred to item forecasts in the forecasting component group or the master demand schedule and then transferred to the material plan.

  1. API Programs

    M3 BE and DMP communicate through application programming interfaces (API). The following API programs are used:

    Program Description
    (OSS401MI) Transfers datasets and data to and from DMP
    (OSS402MI) Transfers accumulator fields to and from DMP
    (OSS406MI) Transfers transaction types to DMP
    (OSS420MI) Returns sales budgets from DMP

  2. Integration with Supply Chain Planner

    M3 Demand Planner has an interface to M3 Supply Chain Planner (SCP) where a supply chain planning simulation can be performed on a demand plan proposal. After planning simulation in M3 SCP, suggested adjustments to the demand plan proposal are saved in a database file and exported back to DMP. Corresponding sourcing rules or a master demand schedule can be transferred directly from M3 SCP to M3 BE.

Related topics